A consistent and distinguishing feature of HIV-associated neurologic disease, particularly dementia and vacuolar myelopathy, is the presence of increased numbers of activated macrophages and microglia in central nervous system (CNS) tissue. These cells are known to produce a variety of potentially neurotoxic substances; elevated CNS levels of TNF-alpha, prostaglandin E2, beta2 macroglobulin, neopterin and quinolinic acid have all been shown to correlated with both the development and severity and HIV-associated dementia. Nevertheless, it is not clear whether these products directly cause damage to neurons or simply are markers of macrophage activation in general. Several studies, however, suggest that macrophage infiltration and activation within the CNS precedes the appearance of dementia. This provides indirect evidence that these immunologic processes are prerequisite events in the development clinically apparent disease. Yet while dementia is a common problem in the late stages of HIV infection, it is not a universal occurrence. Furthermore, the stimuli that cause this local macrophage infiltration and activation over the course of HIV infection in neurologically symptomatic patients remain unknown. In this Project, we seek to build upon our previous work in order to further characterize the processes of macrophage infiltration and activation within the CNS during HIV infection and to determine how they differ in patients who develop HIV-associated dementia from those that do not. Specifically, we aim: 1) To identify the initial patients who develop HIV-associated dementia from those that do not. Specifically, we aim: 1) To identify the initial determinants of macrophage entry into the CNS, 2) To determine the role that matrix metalloproteinases play in the ability of macrophages to extravasate into the CNS, 3) To study determinants of macrophage numbers once they have entered the CNS, and 4) To determine what controls macrophage activation within the CNS.